基于深度学习的阵列信号测向方法综述

A survey of array signal direction finding methods based on deep learning

阵列信号测向在雷达、通信、无线电监管、电子对抗等领域均有着广泛的应用。近年来,基于深度学习的阵列信号测向方法以其在测向性能和测向速度方面的优势,日益成为研究的焦点。系统地梳理了基于深度学习的阵列信号测向技术,首先介绍了阵列观测模型和深度学习相关基础,随后详细阐述了基于深度学习的阵列信号测向基本框架,并对基本框架中数据预处理、深度神经网络结构、数据后处理3部分进行了详细分析和阐述,最后总结展望了该技术未来的发展方向。

Array signal direction finding has a wide range of applications in radar,communications,radio reg⁃ulation,electronic countermeasures and other fields.In recent years,the array signal direction finding method based on deep learning has increasingly become the focus of research due to its advantages in direction finding per⁃formance and speed.The array signal direction finding technology based on deep learning is systematically sorted out.Firstly the array observation model and the relevant foundation of deep learning is introduced,then the basic framework of array signal direction finding based on deep learning is elaborated,and finally the future develop⁃ment direction of this technology is prospected.